The rapid advancement of Artificial Intelligence (AI) has accelerated the development of personalised healthcare. However, the clinical adoption of deep learning remains constrained by a persistent âtrust gapâ surrounding model transparency, security, and data privacy. This talk presents a research vision for building trustworthy human-centred AI in computer audition and biosignal processing by integrating signal processing, machine learning, and healthcare. This vision is structured around three interconnected pillars: Sensing, Understanding, and Empowerment. The Sensing pillar focuses on enabling machines to perceive clinically relevant information from multimodal biosignals, transforming body sounds and physiological signals into non-invasive, accessible, and cost-effective windows into human health and well-being. The Understanding pillar addresses the black-box nature of modern AI systems, with the goal of ensuring transparency, efficiency, robustness, and security in clinical deployment. This is achieved through the development of explainable AI methods, knowledge distillation techniques, and defences against adversarial attacks, fostering AI systems that clinicians can trust and interpret. The Empowerment pillar seeks to restore natural communication for individuals with speech impairments, such as laryngectomy patients. By developing Silent Speech Interfaces (SSIs) that translate facial muscle activity (EMG) directly into audible speech, this research advances speech intelligibility, naturalness, real-time causal architectures, and multi-speaker communication scenarios. Collectively, these research directions aim to establish a foundation of trustworthy, human-centred AI that is not only accurate and secure, but also empathetic, accessible, and impactful, ultimately enhancing healthcare, communication, and quality of life.
论文检索
输入标题、作者或关键词，从 2,909 篇学术成果中精准定位
Recent developments in brain recording are driving a demand for machine learning tools capable of decoding the latent structure of large populations of neurons. In this paper, we provide a comprehensive survey that outlines the trajectory of Latent Variable Models (LVMs) from early state-space models to more recent deep generative models. We organize the literature into three closely related domains: (1) Single-Region Latent Dynamics, which includes models such as linear dynamical systems to more complex dynamics represented by Recurrent Neural Networks (RNNs) and Neural Ordinary Differential Equations (ODEs); (2) Multi-Region Communication, which employs probabilistic as well as subspace methods to study how information is transfered across different brain areas considering synaptic propagation delays and network connectivity; and (3) Behavior-Aligned Modeling, which seeks to disentangle neural activity related to task performance from other internal states via supervised or contrastive learning. Finally, we conclude and discuss benchmarks, evaluation criteria, and open challenges, such as the ability to identify causal links or directionality of communication, to facilitate future research for bridging interpretable brain dynamics with reliable neural decoding.
Most of the widely used estimators of the average treatment effect (ATE) in causal inference rely on the assumptions of unconfoundedness and overlap. Unconfoundedness requires that the observed covariates account for all correlations between the outcome and treatment. Overlap requires the existence of randomness in treatment decisions for all individuals. Nevertheless, many types of studies frequently violate unconfoundedness or overlap; for instance, observational studies with deterministic treatment decisions, popularly known as Regression Discontinuity designs, violate overlap. In this paper, we initiate the study of general conditions that enable the identification of the average treatment effect, extending beyond unconfoundedness and overlap. In particular, following the paradigm of statistical learning theory, we provide an interpretable condition that is sufficient and necessary for the identification of ATE. Moreover, this condition can be used to characterize other treatment effects, such as the average treatment effect on the treated (ATT), as well. To illustrate the utility of our condition, we present several well-studied scenarios where our condition is satisfied and, hence, we prove that ATE can be identified in regimes that prior works could not capture. For example, under mild assumptions on the data distributions, this holds for the models proposed by Tan (2006) and Rosenbaum (2002), and the Regression Discontinuity design model introduced by Thistlethwaite and Campbell (1960). For each of these scenarios, we also show that, under natural additional assumptions, ATE can be estimated from finite samples. We believe these findings open new avenues for bridging learning-theoretic insights and causal inference methodologies, particularly in observational studies with complex treatment mechanisms.
Deep neural networks often rely on spurious correlations, i.e., superficial associations between non-causal features and prediction targets. While yielding high overall accuracy during training, such reliance degrades generalization on minority groups where these correlations break down. Existing methods mitigate this issue by using external group annotations or auxiliary deterministic models, but group annotations are costly to obtain and deterministic auxiliaries may fail to capture the full spectrum of biases learned by the model. We propose Evidential Alignment, a framework that leverages uncertainty quantification to identify and suppress spurious correlations without requiring group annotations. By transforming a biased ERM model from first-order to second-order risk minimization with a Dirichlet distribution and applying an evidential calibration step that reweighs samples by their epistemic uncertainty, Evidential Alignment debiases the model while preserving core features. We provide theoretical guarantees and demonstrate strong worst-group accuracy across diverse architectures and modalities.
With the growing adoption of artificial intelligence in high-stakes decision-making domains, identifying the causes of outcomes--particularly failures--and determining who is responsible has become a critical concern. In this work, we investigate how well formal definitions of responsibility attribution, grounded in the framework of actual causality, align with human judgments of responsibility. To this end, we conduct a large-scale survey to elicit human judgments of responsibility in multi-agent sequential decision-making scenarios, using a modified version of the card game Goofspiel. We evaluate different responsibility attribution methods, assessing their alignment with human judgments about responsibility, and identifying factors that significantly shape responsibility judgments. While no single responsibility attribution method consistently aligns with human responses, our findings highlight key factors that influence human responsibility judgments, including agent-specific biases and the amount of information available to agents during decision-making.
Brain connectivity analysis is a fundamental tool for identifying biomarkers and understanding of neurological disorders. Most existing approaches employ graph transformers over undirected functional connectivity networks, which are typically estimated using correlation statistics. Although effective for capturing statistical associations, these models do not represent directed interactions between brain regions that arise from causal relationships. As a result, direction-specific disease mechanisms are not explicitly modeled, and interpretability is often limited. To address this gap, we present BrainCGT, a brain graph transformer designed to model causal connectivity inferred from fMRI time-series data. In this framework, brain networks are modeled as directed graphs with a modular organization, where nodes correspond to individual brain regions and directed edges reflect causal flow of information between them. Direction-aware node representations together with direction-biased attention mechanisms allow the model to capture asymmetric interactions across regions. Experimental results on three large-scale fMRI datasets demonstrate that BrainCGT achieves consistently better performance than existing graph-based methods for neurological disorder classification. In addition, examination of the learned attention structures shows correspondence with established neurobiological pathways, suggesting improved interpretability. These results highlight the importance of incorporating causal directionality into brain graph transformer architectures for robust and interpretable neuroimaging analysis.
Anterior cruciate ligament (ACL) injury is one of the most common and serious injuries in sports, particularly among recreational skiers. Research shows that structured technique awareness and continuous feedback can significantly reduce the risk of such injuries, yet access to professional instructors is limited to wealthy athletes who can afford continuous private coaching, creating a harmful inequity in injury prevention. This gap can be mitigated by automating the real-time analysis of skiing techniques available to the wider recreational skiing community. The approach relies exclusively on inertial sensors embedded in standard smartphones, eliminating the need for specialized equipment and enabling broad social scalability. To support immediate feedback, the system operates causally, producing predictions based solely on past observations. The work is conducted in cooperation with professional ski instructors, ensuring that problem formulation, data annotation, and result evaluation reflect real-world coaching practices and injury prevention needs. The model is evaluated using Leave-One-Subject-Out validation on a public, in-the-wild dataset, demonstrating robust generalization across skiers, achieving an average directional accuracy of 89.8%, while maintaining extremely low inference latency suitable for on-device mobile deployment. This work outlines a practical pathway to democratizing injury prevention in recreational sports.
Effective flood risk management relies on accurate forecasting, yet the ”black box” nature of state-of-the-art Deep Learning models creates a barrier to trust and accountability in high-stakes public safety decisions. While existing Explainable AI (XAI) methods offer local attributions, they fail to provide the verifiable, operationally meaningful causal narratives required by disaster response authorities. To address this societal challenge, we propose Context-Aware Concept Distillation (CACD), a framework developed in collaboration with domain experts to distill opaque LSTMs into interpretable, hydrology-aware surrogate models. We introduce an unsupervised pipeline to discover a ”Hydrological Language” and a Residual Hypernetwork that dynamically modulates these concepts based on static basin characteristics. Evaluated on 5,203 basins globally, our model achieves high fidelity (Median NSE 0.70), significantly outperforming black-box baselines (e.g., Multi Layer Perceptrons) on unseen future data. By demonstrating that human-interpretable concepts are sufficient to reconstruct flood dynamics, this work balances AI accuracy with the transparency required for responsible environmental decision-making.
Vision-language-action models (VLAs) often achieve high performance on demonstrated tasks but struggle significantly when required to extrapolate, recombining skills used in different tasks in novel ways. For instance, VLAs might successfully put the cream cheese in the bowl and put the bowl on top of the cabinet, yet still fail to put the cream cheese on top of the cabinet. This motivates us to investigate whether VLAs merely overfit to demonstrated tasks or still hold the potential to extrapolate. Our study uses text latent as the ingredient; it is a task-specific vector derived from the models’ hidden states. It thus encodes semantics necessary for completing a task and can be used to reconstruct the associated task behavior by writing it to the model’s residual stream. Furthermore, we find that skills used in distinct tasks can be combined to produce novel behaviors by blending their respective text latent. Applying this to π0, we increase its success rate from 9% to 83% on the proposed libero-ood benchmark, which features 20 tasks extrapolated from standard LIBERO tasks. This reveals that the skill representations encoded in text-latent are individual yet composable, while π0 fails to autonomously combine these representations for extrapolation. This also validates the design of libero-ood; it comprises tasks that the model fails, yet should be able to complete. We then tested other VLAs on libero-ood, and none of them achieved a success rate higher than 21%. Further analysis reveals VLAs share a common pattern to exhibit spatial overfitting, associating object names with where the object is spatially located in the demonstrated scene rather than achieving true object and goal understanding.
While Large Language Models (LLMs) offer promise in scientific discovery, leveraging LLMs to drive biomedical research requires the scientific discovery process to be performed in combination with cutting-edge biomedical research and rigorous mechanistic causal chains. As such, both current Retrieval-augmented generation methods lacking causal reasoning capabilities, and the static traditional knowledge graphs failing to reflect evolving scientific knowledge, present obstacles to utilizing LLMs as scientific discovery tools. In response to these ongoing challenges, we present THGAgents. THGAgents utilize collaborative and dynamically updating agents to build a Traceable Causal Knowledge Graph, which serves as the foundation for the evidence-based knowledge structure. Crucially, we employ an LLM-driven heuristic search algorithm to traverse the complex network, balancing both novelty and rigor to deduce strict, evidence-based mechanistic causal chains. Additionally, THGAgents utilize a generator-critic loop to support hypothesis refinement. In experimental benchmarks across both cancer systems and neuroscience, THGAgents achieved up to a 0.80 hit rate in predicting validated scientific discoveries, providing an almost 9.5% increase in hypothesis quality scores versus current state-of-the-art systems, and decreasing the mechanistic hallucination rate to 1.12%. Our code is available at https://github.com/yangCode-res/THGAgents/.
In multi-turn interactions, large language models (LLMs) often exhibit a persistent influence from prior turns, even after an explicit topic switch. This behavior, which we term semantic inertia, can cause responses to deviate from the expected output distribution for an independent task, undermining task isolation and reliability. This paper introduces a rigorous experimental framework to systematically characterize the nature, form, and dynamics of semantic inertia. We propose an operational definition and a causal-contrastive method that isolates semantic carryover from confounding factors like context length. Through a series of experiments on five leading LLMs, we (i) confirm the existence of semantic inertia and identify its boundary conditions; (ii) model its decay over the course of generation, revealing a characteristic timescale and a heavy-tailed distribution; (iii) decompose its effects on three distinct channels—factual accuracy, structural integrity, and stylistic expression; and (iv) probe its controllability using prompt-based interventions. Our key findings show that inertia is not a simple length effect but an intrinsic dynamic, strongest in the initial part of a generation and decaying over a timescale of approximately 100-200 tokens. Its impact is most pronounced as a stylistic residue, while its effect on factual correctness is weaker and highly dependent on the task and domain switch. Crucially, we find that prompt-level ``reset'' instructions are unreliable and often counter-productive, while conflicting constraints consistently amplify, rather than resolve, output deviation. These results suggest that governing semantic inertia requires system-level state management mechanisms rather than relying on prompt engineering alone.
Identifying meaningful latent representations within diffusion models remains a challenging problem for causal approaches. We propose Causal Manifold Transport Diffusion Model (CMT-Diff), a framework that operationalizes causal actions as geometric transformations. By adopting the perspective of backtracking counterfactuals, we formulate the generative process as a composite diffeomorphism that couples the Probability Flow ODE with a Continuous Normalizing Flow. This mapping constructs an exogenous manifold where causal factors align with coordinate variations. Within this geometry, we derive Causal Manifold Transport (CMT) to realize interventions as linear vector translations along factor-aligned directions. We establish theoretical identifiability guarantees and demonstrate that our approach facilitates controllable generation by capturing the underlying causal manifold.
With the increasing demand of privacy protection, Machine Unlearning (MU) appears to remove private data from an already trained model without retraining from scratch. Most current works suffer from overly unlearning (low fidelity) or incomplete unlearning (low effectiveness). To identify the issues behind, we conduct causal analysis to obtain a resolvable route, i.e., disentangling shared knowledge into attribute-level semantics to remove it as the confounder. We further perform MU loss analysis to reformulate it as balanced form of constraints, thus guaranteeing high fidelity and effectiveness. Based on theoretical analysis, we propose disentangled knowledge forgetting constrained by the reformulated MU loss, which disentangles knowledge with variational auto-encoder and refines knowledge with counterfactual inference. Extensive experimental results demonstrate that our method achieves state-of-the-art performance.
Continuous-Time Dynamic Graphs (CTDGs) enable fine-grained modeling of evolving relational systems. However, most existing CTDG representation learning methods are tailored to in-distribution settings and exhibit limited robustness under out-of-distribution (OOD) shifts. Although recent causal approaches learn invariant representations via interventions, they are primarily designed for static or discrete-time graphs and become computationally prohibitive for CTDGs due to the combinatorial explosion of structural and temporal variations. To address these challenges, we propose CIR, a framework grounded in a novel structural causal model termed the ICCM. To avoid exhaustive interventions, we leverage the Normalized Weighted Geometric Mean (NWGM) to efficiently approximate interventional predictions. We further instantiate ICCM within a practical deep learning architecture that jointly captures invariant structural and temporal patterns through dedicated subgraph extractors, and maintains an environment memory bank to model distributional shifts across evolving contexts. Extensive experiments demonstrate that CIR consistently outperforms existing methods under diverse OOD scenarios.
Fairness-aware graph learning has become increasingly important amid growing concerns about algorithmic bias in networked data. Among existing approaches, counterfactual fairness is particularly appealing as it seeks to eliminate unfairness at its causal origin by ensuring that predictions remain invariant in counterfactual worlds where sensitive attributes are altered. However, most existing methods assume that all observed variables are directly influenced by sensitive attributes, an overly strong and often unrealistic assumption in real-world graphs. To address this limitation, we propose Graph Counterfactual Fairness (GCFair), a novel framework that achieves counterfactual fairness by explicitly identifying and disentangling the subsets of node features and graph structures genuinely affected by sensitive attributes. This principled joint disentanglement enables the generation of authentic counterfactual instances that selectively modify only sensitive-related information while preserving all sensitive-irrelevant factors. Extensive experiments show that GCFair effectively mitigates bias and outperforms state-of-the-art fairness methods in both counterfactual fairness and predictive accuracy.
While Vision-Language Models (VLMs) have shown promise in textual understanding, they face significant challenges when handling long context and complex reasoning tasks. In this paper, we dissect the internal mechanisms governing long-context processing in VLMs to understand their performance bottlenecks. Through the lens of attention analysis, we identify specific Visual Evidence Retrieval (VER) Heads —a sparse, dynamic set of attention heads critical for locating visual cues during reasoning, distinct from static OCR heads. We demonstrate that these heads are causal to model performance; masking them leads to significant degradation. Leveraging this discovery, we propose VERA (Visual Evidence Retrieval Augmentation), a training-free framework that detects model uncertainty (i.e., entropy) to trigger the explicit verbalization of visual evidence attended by VER heads. Comprehensive experiments demonstrate that VERA significantly improves long-context understanding of open-source VLMs: it yields an average relative improvement of 21.3% on Qwen3-VL-8B-Instruct and 20.1% on GLM-4.1V-Thinking across five benchmarks.
Probabilities of causation (PoCs) are fundamental quantities for counterfactual analysis and personalized decision making. However, existing analytical results are largely confined to binary settings. This paper extends PoCs to multi-valued treatments and outcomes by deriving closed form bounds for a representative family of discrete PoCs within Structural Causal Models, using standard experimental and observational distributions. We introduce the notion of equivalence classes of PoCs, which reduces arbitrary discrete PoCs to this family, and establish a replaceability principle that transfers bounds across value permutations. For the resulting bounds, we prove soundness in all dimensions and empirically verify tightness in low dimensional cases via Balke's linear programming method; we further conjecture that this tightness extends to all dimensions. Simulations indicate that our closed form bounds consistently tighten recent recursive bounds while remaining simpler to compute. Finally, we illustrate the practical relevance of our results through toy examples.
We introduce missingness-MDPs (miss-MDPs), a novel subclass of partially observable Markov decision processes (POMDPs) that incorporates the theory of missing data. A miss-MDP is a POMDP whose observation function is a missingness function, specifying the probability that individual state features are missing (i.e., unobserved) at a time step. The literature distinguishes three canonical missingness types: (1) missing completely at random (MCAR), (2) missing at random (MAR), and (3) missing not at random (MNAR). The problem is to compute near-optimal policies for a miss-MDP with an unknown missingness function, given a dataset of action-observation histories. Achieving such optimality guarantees for policies requires learning the missingness function from data, which is infeasible for general POMDPs. To overcome this challenge, we exploit the structural properties of different missingness types to derive probably approximately correct (PAC) algorithms for learning the missingness function. These algorithms yield an approximate but fully specified miss-MDP that we solve using off-the-shelf planning methods. We prove that, with high probability, the resulting policies are ε-optimal in the true miss-MDP. Empirical results confirm the theory and demonstrate superior performance of our approach over two model-free methods.
Detecting hidden confounding is crucial for reliable causal analysis from observational data, directly determining which downstream causal inference method to be deployed. Inspired by the theory of higher-order regression, recent sample-efficient hypothesis testing strategies overcome the restrictive requirement of multiple heterogeneous data environment. Despite their progress on single-environment confounder detection, such methods suffer from intrinsic flaws that the structural functions of the causal models should be specified in prior (linear or specific kernel functions). By contrast, real-world data acquisition exhibits diverse, unknown forms of structural functions, imposing an important but challenging gap between theories of higher-order regressions to practical confounder detection. In this paper, we contribute a Bi-level Kernel Confounder Detection (BiKCD) framework by learning adaptive kernelized structural space of structural functions. Subsequently, our BiKCD constructs hypothesis testing by comparing coefficients from the higher-order regression and the classical ordinary least squares in learned kernelized space. Finally, the hypothesis is calibrated to ensure valid inference under adaptivity. Theoretically, we establish an oracle-type risk bound for the selected structural space over a candidate kernel family, with the Type-I error control for the downstream test. Extensive experiments on synthetic and real-world datasets demonstrate the effectiveness of the proposed BiKCD.
Machine unlearning (MUL) aims to remove the influence of specific data from a trained model for data privacy and model adaptability. Existing MUL methods mostly assume the internal parameters and the training data of the target model are accessible. Nevertheless, in most practical scenarios, the model provider (MP) and the service operator (SO) are different entities with unequal model access privileges. The MP provides the model, while the SO can only access the model via APIs when handling unlearning requests. Under such an asymmetric access constraint, we propose One-Turn Knockout (OTK), a novel traceable and editable MUL framework based on a model-agnostic and editable proxy. Specifically, OTK first compresses the representation space of the target model into a discrete proxy based on codebook, with merely one pass post-training. Each data sample is recorded in the proxy space as a distribution over the codebook tokens, and its contribution to the model prediction can be cumulatively estimated via additive token statistics. Based on the traceable and editable proxy, the SO can instantly handle unlearning requests by (i) estimating the token distribution of the forgotten data, (ii) identifying the causal tokens, and (iii) erasing their contributions without the access to the model parameters and training data. Extensive experiments on multiple datasets and tasks show that OTK consistently outperforms state-of-the-art unlearning methods.